CN1879091A - Method and device for persistent-memory management - Google Patents
Method and device for persistent-memory management Download PDFInfo
- Publication number
- CN1879091A CN1879091A CNA2003801028714A CN200380102871A CN1879091A CN 1879091 A CN1879091 A CN 1879091A CN A2003801028714 A CNA2003801028714 A CN A2003801028714A CN 200380102871 A CN200380102871 A CN 200380102871A CN 1879091 A CN1879091 A CN 1879091A
- Authority
- CN
- China
- Prior art keywords
- file
- data
- file system
- storage
- memory
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F12/00—Accessing, addressing or allocating within memory systems or architectures
- G06F12/02—Addressing or allocation; Relocation
- G06F12/08—Addressing or allocation; Relocation in hierarchically structured memory systems, e.g. virtual memory systems
- G06F12/0802—Addressing of a memory level in which the access to the desired data or data block requires associative addressing means, e.g. caches
- G06F12/0866—Addressing of a memory level in which the access to the desired data or data block requires associative addressing means, e.g. caches for peripheral storage systems, e.g. disk cache
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F12/00—Accessing, addressing or allocating within memory systems or architectures
- G06F12/02—Addressing or allocation; Relocation
- G06F12/0223—User address space allocation, e.g. contiguous or non contiguous base addressing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/10—File systems; File servers
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F2003/0697—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers device management, e.g. handlers, drivers, I/O schedulers
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The present invention relates to a method for managing memory space of a persistent-memory device and to a memory management device. The memory management method of the invention comprises a step of allocating (S14) at least one first part of said memory space to a file system (74) upon request from said file system (74) or from an application (70). The method and the device of the present invention enable a dynamical allocation of persistent-memory space to a file system. This way, the memory space of a persistent memory is effectively used also for write-caching. At the same time, write-caching and storing steps can be accelerated.
Description
The present invention relates to a kind of memory management apparatus and a kind of method that is used to manage the storage space of persistent memory device.The invention further relates to a kind of method that is used at the persistent memory device write-in cache.It also further relates to and a kind ofly is used to use persistent memory device being saved in method on the disk by using the data of handling.It also relates to a kind of file system device, application apparatus and data handling system.
Finish the persistent data storage hereof.File is the limited linear sequence of data, for example by using the data of creating and using.In the notebook data disposal system, file is stored in the secondary storage medium enduringly, it has the write access time of comparing slow with data processing speed.The example of this secondary storage medium is tape or rotation hard disk, below also is called disk, and they all have magnetically can write and readable coating.Disk comprises equal-sized data block, is called disk block or sector.In the data handling system of picture computing machine and so on, the data of file system management persistent storage.It is for file allocation disk block chain and write these pieces.
In order to improve processing speed, being written to fast-access storage by the data structure of using processing or being loaded into the fast-access storage from using the data structure of handling from disk, be also referred to as random-access memory (ram), below also be called working storage.Distribute and the storage space of management in working storage by memory manager program.
Term application as used herein relates to process or the equipment that is suitable for handling memory data.Data manipulation for example comprises to storer and writes data, stores on the secondary storage medium from memory erase data, conversion memory data or memory data.The form that employing is loaded into the one or more executable files in the storer is embodied as equipment to application, and storer is connected to for example central processing unit in computing machine.Also can be embodied as for example integrated circuit of special IC (ASIC) and so on to application, it is suitable for carrying out mentioned process.File system and memory manager program are considered in attention under the situation of specialized application.
The structure of operational data generally is the many storage blocks that connected by pointer in working storage.For example, the operational data structure in storer can be in continuous blocks or in interspersing among the piece of working storage, with pointer block chaining be got up in both cases.In order to programme simple and easy use and speed preferred in storer the structure of operational data.It can have obviously different with serial representation hereof, and it needs usually to observe given propriety standard usually.
Therefore, be data structure mapping file, use handle data structures relating to according to predetermined conversion (serialization) process in the file of the data storage of coming self-application to the disk.The serialization process depends on the application-specific of creating file, and is therefore carried out by application itself.
Be used for be written to from the data of working storage the file on the disk to access time of disk in the millisecond scope.This compared and will grow with the processing time of being expended by microprocessor, microcontroller, digital signal processor (DSP) or the like.Therefore, disk access is the bottleneck in data processing stream.
In order to overcome to disk ablation process shortcoming slowly, the general use has the cache memory that writes that writes the fast access time and comes high-speed cache will be stored in data on the disk.Be cached to refer in the notebook data disposal system and at first the operational data structure copied to fast access and write cache memory.The cache memory that writes that below also is called cache memory separates with working storage and preserved data before data are write disk in storer.Therefore preserving file relates to operational data being transformed to serialized data and this serialized data being written to from working storage and writes cache memory, this finishes by should being used for, and at last serialized data from writing the disk space that cache memory is written to distribution, this is finished by file system.
The advantage that writes cache memory be it allow to be applied in needn't the situation of Interrupt Process under initialization in the persistent storage of file on the disk up to finishing on disk storing process slowly.In case serialized data is safe in cache memory, file system just sends acknowledge message to using.After confirmation of receipt message, use and prepare to continue to handle.Manage persistent storage on disk by file system then.Application can be finished or work at that time or write file to disk.
Known two kinds of perishability cache memories: piece level cache memory and file-level cache memory.In piece level cache memory, preserve file according to the serialized data minor structure, minor structure is suitable for the size of disk block and is assigned on disk specific.On the other hand, in file-level cache memory, file is saved as serialized data with the similar minor structure of piece.Postpone from the conversion of file structure by file system, in fact need to write data into disk up to it to the disk storage block structure.Thereby, in file-level cache memory, before the actual process that writes of disk, do not have to specific disk block distribute data in initialization.File system only guarantees that the somewhere that can be stored in the file data in writing cache memory on the disk gets final product.
Serialization in working storage data and serialized data storage spent the considerable time to disk, use during this period and can not confirm successfully to store data.In addition, the existing memory devices such as dynamic RAM (DRAM) equipment depends on non-persistent data storage.Data in the perishability storer can be lost in the system failure.Therefore, write in the cache memory and data can not be preserved for a long time in perishability, this is too high because of the risk of losing significant data owing to the system failure.Therefore, be used to store by using the process of handling data, that described complexity is time-consuming again and must usually be performed so that prevent loss of data.This slows down file system performance.
At present, persistence and simultaneously working storage equipment-such as based on RAM (MRAM) equipment of magnetic resistance-from the whole world research and development activity of concentrating, occur fast.Even persistent memory device is gone back recording storage data therein after power down.Use this advantage of non-volatile storage, can attempt to improve the performance of file system.
People such as Wang have described a kind of disk/persistent RAM mixed file system in following document, An-I A.Wang, P.Reiher, G.J.Popek, G.H.Kuenning, " Conquest:Better Performance Through a Disk/Persistent-RAM Hybrid FileSystem " can find on http://www.lasr.cs.ucla.edu/awang/papers/usenix2002/camera2.htm.Their file system is to be based upon on the following hypothesis, and non-volatile storage is available on a large scale in the GB scope.The size of file is depended in the use that is used for file storing space or disk space according to people's such as Wang file system to a great extent.It is at little this basis of file that people's such as Wang file system is based on most of file access steps.In addition, the major part of little file composition file in file system.Most of space of fast access non-volatile storage is for the small documents below 1 megabyte, metadata, can carries out in persistence RAM and the storehouse of sharing keeps.Metadata is the data that comprise about file data information.The example of metadata is the i node.The i node is retained in the management information and the address that is included in the data block of the file data on the disk about file of disk.Therefore, the big file of slowly visiting based on disk is limited in smaller number of events.Like this, according to people such as Wang, aspect access speed, improved the performance of whole file system.
Yet modern Application is owing to abundant auxiliary storage space is that disk space trends towards generating increasing file.Therefore, make and widely applying data security to keep slow process in this document system on the disk.In addition, exist the hypothesis of abundant quick and durable property storage space not to be suitable at present in commercially available most of data handling systems.
People such as Miller have advised that a kind of persistence RAM that uses is as the file system (E.L.Miller that is used for the permanent memory medium of file system data and metadata, S.A.Brandt, D.D E.Long, " HeRMES:High-Performance Reliable MRAM-EnabledStorage ", according to 8th symposial of operating system-HOT0S VIII about hot topic, Schloss Elmau, Germany, May calendar year 2001).Also MRAM as write buffer in this document system.This allows to postpone by wish not have to the ablation process of disk owing to the system failure and the risk of obliterated data.Yet the document does not have the connotation of public use MRAM as write buffer on persistent-memory management.
Therefore an object of the present invention is to provide a kind of memory management apparatus and a kind of method that is used to manage the storage space of persistent memory device, it allows persistent memory devices with for writing cache memory.
Another object of the present invention provides a kind of method that is used for write-in cache, and it has improved the speed that makes application data safety.
Further aim of the present invention provides a kind of method that is used for preserving to disk the data of being handled by application, and it has improved the performance of file system on speed.
Further aim of the present invention provides the file system of the reliability with increase, and it allows to use the bottleneck of avoiding to the secondary storage medium ablation process.
Further aim of the present invention provides a kind of application apparatus, and it supports storage management method of the present invention and equipment.
Further aim of the present invention provides a kind of data handling system with file system performance of enhancing.
Below, in order to be that the understanding of the present invention is easier, before will explaining aspect the equipment of the present invention, will at first explain the present invention with respect to its method aspect.
Method with respect to the storage space that is used to manage persistent memory device, this purpose is that the method by the storage space that is used to manage persistent memory device realizes, comprise according to request, at least one first of described storage space is distributed to the step of described file system from file system or application.
According to method of the present invention, at least one part of the storage space of persistent memory devices is distributed to file system.Memory allocation is that described part storage space is such as many according to the process of request distribution portion storage space.Is uncommon to file system memory allocated space in the known storage management method of prior art, and this is because the file system that prior art is considered only is in charge of the persistence auxiliary storage space of picture disk space and so on.Yet, except that disk space, can distribute the part storage space of persistent memory device for file system according to method of the present invention.Thereby after allocation step, the storage space under file system is responsible for not only comprises disk space, also comprises the part storage space of persistent memory device.
By coming to file system distribution portion storage space according to the inventive method, not only corresponding storage space is included in data in this part storage space in addition and is also handed to file system and be responsible for.After executing allocation step, application can not visit again the data of having been given respectively in the storage space of using when the time point that distributes to file system requests.Become more clear at this point, allocation step is formed for making the basis of the new mode of operational data safety in persistent memory device.Use method of the present invention, can be used as working storage and write cache memory by the persistent memory devices.
According to the inventive method, according to from file system itself or come the request of self-application to carry out allocation step.Thereby, only when requiring, give file system memory allocation.The persistent-memory space that does not in advance keep for file system.This allows effectively to utilize the fast access non-volatile storage, when especially having limited amount sort memory.Be thought of as any process of carrying out by processing unit, processing unit such as general central processing unit (CPU), special IC (ASIC), digital signal processor (DSP) or ASIP (ASIP) according to application of the present invention.Process can be the subprocess of the more comprehensive application system such as the word processing application system.Can ad hoc represent to use by the executable file that is loaded into working storage.The working storage that comprises executable file can be a non-volatile storage, for example identical non-volatile storage managed of the inventive method, or perishability storer.
It is the file system memory space dynamic allocation in persistent memory device that storage management method therefore of the present invention allows.It is not the file system priori reservation of memory space.The current state that depends on storer for the storage space of file system distribution.
Thereby, depend on storer and the current state that makes memory-aided data handling system, whole storage spaces can be controlled by memory management unit, short ofly just can carry out conventional storage means according to principle known in the state of the art to the described memory management unit in file system requests memory allocated space.In case when being used to the request of file system distribution portion storage space, reduced by this part by the available storage space of this conventional memory management by the memory management unit reception.
Thereby, the storage space of method formation persistent memory devices of the present invention uses the method as fast access working storage and fast access file system memory, and described fast access file system memory writes cache memory such as file system.Can be in time identical memory allocation be used to one a moment, it can be distributed to file system, and afterwards it was distributed to identical again or Another Application in the moment after a while.
Certainly, can use method of the present invention to manage the storage space of a plurality of persistent memory devices.The storage space that will be managed be each persistent memory device storage space and and it can be treated (addressing) as a logical memory space.
In an allocation step, there is the above storage space of a part for the persistent memory device of file system distribution.Generalized case is the operational data that has two or more groups in the working storage that is formed by persistent memory device, and it is by identical application parallel processing.For example, just before finishing application, if the operational data that application request makes two groups safety all, allocation step can comprise the storage space that comprises two groups of operational datas of these two parts.Certainly, if there is no speed issue can also make the data security of these groups in succession.
The preset upper limit that storage space can be arranged, such as the megabyte number that can distribute to file system, in order to avoid following situation occurs, become rareness and system performance of the storage space that wherein needs to become working storage becomes slow.
In first preferred embodiment of storage management method of the present invention, allocation step comprises the step of obstruction to storage space first write access.By this step, the data that are included in this part of storage space are safe, promptly prevent from self-application or from any write access of file system.
In a further embodiment, allocation step comprises the ability of dividing the storage space read access of pairing first to file system.Like this, the data of being responsible for being included in the storage space first are fully handed to file system.Any read access to data after this step must be by file system management.This wherein writes disk to operational data corresponding to the situation that is different from the routine data disposal system.Yet, in the present embodiment, also be not written into disk and safety still under the situation of the system failure even can data.The application of these data of request visit does not note being stored in usually the difference of data on the disk.The present embodiment permission for example under the situation of the system failure, is applied in and restarts afterwards via file system fast access data.Thereby present embodiment is useful for the write-in cache in non-volatile storage.After the data security, do not need to carry out immediately the step that writes disk making according to allocation step in non-volatile storage.The step that writes disk can be by the stage after putting off until and without any the risk of loss of data.Application guarantees after the allocation step successful execution of present embodiment, just as made operational data safety immediately in the disk access step of routine.
Another preferred embodiment of storage management method of the present invention comprises removes the step of distributing to the first of storage space from memory manager program.In this embodiment, file system " is given back " storage space that those have been assigned to file system to memory manager program.Like this, storage space can freely be used as working storage once more.It also can be assigned to Another Application.
Preferably, allocation step and/or releasing allocation step comprise address or the address realm that sends definition storage space first.Can be by the first and second address definition address realms, the storage unit separately of persistent memory device is all represented in each address.So address realm will comprise all storage unit of the address that has between first and second addresses.In allocation step, transmission is from the memory manager program to the file system.Therefore file system can be included in the address realm that receives in its file allocation system, such as file allocation table or high-speed buffer storage data allocation table independently.Preferably, memory manager program is labeled as the address realm that is sent and distributes to file system in order to avoid to the secondary distribution of the mistake of same storage space.In redistributing step, transmission is from the file system to the memory manager program.File system will be eliminating from the address realm of its allocation table or it being labeled as inaccessible.Memory manager program will be labeled as the address realm that is sent can freely to reuse, promptly to the new distribution of file system or application.
In further preferred embodiment, described first for the described storage space of given situation carries out the releasing allocation step, and described situation is included in first data in this part of storage space are stored in persistent memory device with the form of file data the second portion of storage space.This embodiment is useful for the process that operational data is saved in file.In case with the form of file structure first data are kept in the persistent memory device and the second portion of storage space and distribute to file system, data are exactly safe so.Therefore, needn't be kept at the original operational data structure of first data in the non-volatile storage again.Can remove the first of allocate memory.Memory manager program can freely reuse this part in new allocation step then.Note, the first of storer has been distributed to this situation of file system before this embodiment only is only applicable to wherein.
In the further embodiment of storage management method of the present invention, to carry out for the described second portion of the storage space of given situation and to remove allocation step, described situation is for writing secondary storage medium to file data.Be that this embodiment is applied to wherein the situation of data storage on disk valuably.After successfully file data being written to the step of disk, also needn't make the file data safety in non-volatile storage.Therefore can also to memory manager program according to aforesaid, remove the second portion in memory allocated space with the duplicate mode of storage space first.
According to a second aspect of the present invention, the said method of storage space of management permanent storage device is used for write-in cache on permanent storage device.The present invention is used for the method for write-in cache by first data of use handling, and described first data are included in the first of storage space of persistent memory device.Method of the present invention comprises the step of coming the execute store management method according to a first aspect of the present invention or its any embodiment.
According to method of the present invention, by distributing the storage space that has been assigned to application to carry out write-in cache to file system.In the context of aforesaid description at storage management method of the present invention, only according to carrying out allocation step from the application of handling first data or the request of file system.This allows to prevent the obstruction that undesired obstruction is used to operational data.
This write-in cache method is introduced the level more taller than file-level high-speed cache.File system is handed in the control of first data, first data are illustrated in the operational data in the first of persistence working storage, thereby create the interim expansion of file system, and its expression file system writes cache memory.In case the allocation step of execute store management method successfully, the data that are included in the non-volatile storage first just can be by write access further.Can write data into the secondary storage medium such as disk subsequently.Can put off until reasonable time to the step that writes disk, but not cause any risk of obliterated data simultaneously owing to the system failure.If necessary, can at first be transformed to file structure to operational data, as in the preferred embodiment hereinafter will as described in.According to this method, compare with known write cache method and to have improved the speed that makes operational data safety in writing cache memory widely.
The preferred embodiment of write cache method of the present invention comprises, the step that acknowledge message is sent to application from file system after described allocation step.Acknowledge message can be sent by memory manager program rather than file system after allocation step in optional embodiment.Acknowledge message tells that application data is safe.For example, if will store operational data, can after confirmation of receipt message, close application immediately so when finishing when using.
In the further preferred embodiment of write cache method of the present invention, first data are included in the copy of the 3rd data in the third part of storage space of persistent memory device.This has described such a case, when using the 3rd data that continue to handle in the non-volatile storage third part, but has to come the order of the current state safety that makes the 3rd data in the file of self-application.In this embodiment, the write cache method comprises, before the execute store management method the step of the 3rd copying data to first storage space.The state of the operational data when therefore first data are illustrated in mentioned order.
This embodiment is particularly important, because the progress of work that it is used to use.This embodiment also clearly show that the advantage of write cache method of the present invention.In case the 3rd copying data to the first of storage space and for the first of storage space has carried out allocation step, use and just can continue to handle the 3rd data.Because the persistent nature of storer, the time after can be carries out first data is write disk.
Preferably, above-mentioned copy step comprise the first of storage space distribute to the copy that is used for the 3rd data application step and this copy of the 3rd data is written to the step of storage space first.
The further preferred embodiment of write cache method of the present invention is applicable to that first data wherein have the situation of the operational data structure that is different from file structure.Not every operational data structure all must be different from file structure.This depends on specific application.Yet according to the standard of application-specific, the operational data in storer first does not show linear structure in general.In this case, the present embodiment of write cache method comprises distributes to the 4th part of storage space that to be used to be suitable for first data conversion be the step of application of the executable file of file data, conversion, executable file is write the tetrameric step of storage space and the step of the 4th part of storage space being distributed to file system.
In this embodiment, application is delivered the routine that is used for operational data is transformed to concrete application file structure to file system.In the replaceable example of this embodiment, routine stores in file system, and use and to transmit quoting of this routine to file system.Can select in the example at these two, routine has the form of executable file, perhaps preferably, has the form of dynamic link library (dll).Should be noted that in general, can use here executable anything, such as just by carrying out the executable code that another executable code is created.Under preferred dll situation, file system can be called the pre-defined function of dll subsequently, transmits pointer so that conversion to storer.The result is in these two embodiment, and it is the task of file system rather than the task of application that operational data is transformed to file data.File system can wait to be transformed during up to reasonable time point or up to the needs conversion.Application needn't be handled conversion and can continue to handle data in the storage space third part.Reduced like this by using time cost, that be used to make operational data safety.Therefore, improved application speed.
In this embodiment, preferably, executable file is write the tetrameric step of storage space by the initialization of application own.More preferably, this is to finish in the order of being given the data security of sening as an envoy to by application or and then.
Hand to file system only needing conversion executable file once.After this, enliven as long as use, file system can remain on transform routine and the 4th storage space that is associated under its task.As selection, can also be transferred to file system to transform routine with each write cache.If persistent-memory space is rare, the example selected of back will be useful so, and when reducing the processing load that causes owing to data storage, it is more preferred further can selecting example.These two kinds can the selection situation under, after shift step or after completing steps, carry out the step of the 4th part of storage space being removed distribution from memory manager program respectively.Like this, memory manager program can be given Another Application memory allocation.
As further replaceable example, it is that time of enlivening is longer that the time ratio that file system can be preserved transform routine is used.By the transform routine in the 4th part of non-volatile storage being remained on the down responsible of file system, improved write cache speed further.This be suitable for only moving one use or the data handling system of application system in particularly useful.On the other hand, only ought have enough persistent-memory space and program is kept at the performance that can not weaken working storage in the storer, but the system of selection talent is useful.
Preferably, by means of executable file is first data conversion that the step of file data comprises the step of the second portion of storage space being distributed to the executable file that is used for file data, create file data by executable file being applied to first data, described file data is written to the second portion of storage space, and the described second portion of described storage space is distributed to described file system.With reference to the explanation of storage management method of the present invention, wherein introduced the second portion and the file data of term storage space.For consistance reuses these terms here.
Following benefit is arranged promptly by the file system initialization shift step: can prevent unnecessary shift step.Use and to have the routine of calculating according to last safe action that makes data security after the time span automatically that pre-determines usually.Do not finish if use, previous copy be not transformed under the situation of file data so, can be overwritten in the operational data copy in the storage space first simply.Like this, utilized the persistent nature of the storage space of the risk that need not bear loss of data again, further reduced and handled load.
According to a third aspect of the present invention, above-mentioned write cache method is used in the context of method the file that is saved in by the data of using processing on secondary storage medium.This preservation document method comprises according to second aspect present invention or its any the foregoing description carries out the write cache method, and the step that file data is written to secondary storage medium.
Preservation document method of the present invention has shown the reliability that improves, and this is that wherein allocation step is carried out during write cache owing to make the data security that will write disk immediately after relating to the allocation step of storage management method.In addition, the use of the write cache method in the preservation document method of the present invention makes application can prevent bottleneck to the secondary storage medium ablation process.Only use non-volatile storage is used the fast access ablation process and in case successfully carried out above-mentioned allocation step with regard to confirmation of receipt message.Manage independently at reasonable time that by file system the actual of disk write, for example using when not needing whole processing power of processing unit.
File data is write disk can be according to known procedures.In a preferred embodiment, data save method comprises, before the said write step, file data is split as file data blocks so that make the structure of the data structure adaptation storage medium on disk in non-volatile storage.Preferably, splitting step comprises the sector of each file data blocks being distributed to uniquely described disk.At last preferably, write step comprises the sector of distributing separately that each file data blocks is written to described disk.
According to a fourth aspect of the present invention, provide a kind of storage management apparatus that is used to manage the storage space of at least one persistent memory device.Storage management apparatus comprises storage allocation location, is used for communicating by letter with at least one application apparatus and at least one first of being used for the storage space of persistent memory devices distributes to application apparatus.According to the present invention, allocation units also are used for communicating by letter with at least one file system device, and are used for according to from described application apparatus or from the request of described file system the first of storage space being distributed to described file system.
Storage management apparatus of the present invention-also be called as sometimes Memory Management Unit-be suitable for carrying out storage management method according to first aspect present invention or its any one embodiment.For this reason, with known storage management apparatus modified in comparison allocation units.For example can provide storage management apparatus of the present invention with the form of the integrated circuit such as ASIC.On the other hand, it can also provide with the executable file form, and this executable file is loaded into the working storage that is connected to processing unit.In particular, the working storage that comprises executable file can be a permanent storage device, and it is under the control of storage management apparatus.
Storage management apparatus of the present invention is in and is suitable for keeping among the preferred embodiment of memory allocation table.Memory allocation table represents the storage space part of the persistent memory device that defines at least one memory allocation described application apparatus or described file system device, described storage address.The situation of distributing to file system or application has been described respectively in the context of storage management method of the present invention.Therefore the appropriate section of this instructions is carried out reference.
According to a fifth aspect of the present invention, a kind of file system device is provided, comprise file allocation unit, be used at current state maintenance documentation allocation table, described file allocation table is distributed at least one file at least one disk space address.File allocation unit is suitable for communicating by letter with the storage management apparatus relevant with persistent memory device.In addition, the file allocation unit address that is used at least one first storage space of persistent memory devices is included in the file allocation table of maintenance.
The file system of this aspect of the present invention is used for working with storage management apparatus as mentioned above, and is used for using at preservation file of the present invention and write cache method.
In a similar fashion, memory management equipment is the same, also can provide file system device with the form of integrated circuit.It can also comprise processor and storer as selection, and realizes file allocation unit with the form that is included at least one second executable file in the described storer.As above-mentioned, can with another equipment shared processing device and storer, another equipment such as storage management apparatus or one or several application apparatus.
According to a sixth aspect of the present invention, a kind of application system is provided, comprise the persistent memory device that is connected to processor and be used for handling Data Management Unit in described permanent storage device data.Data Management Unit is used at least one the 3rd executable file or dll are write described non-volatile storage, so that by carrying out described the 3rd executable file or dll, it is predetermined data-sequence form that described processor is suitable for described data conversion, promptly as creating file data according to operational data will describing in detail further with reference to figure 1 described in the top method of the present invention and as following.Data Management Unit is used for executable file is write persistent memory device, and executable file comprises transform routine, and it is the linear order of data that this transform routine of transform routine is used for the data structure mapping that is included in persistent memory device.As selection, Data Management Unit is used to file system to be provided to such quoting of executable file in the file system or dll that be included in.
Application apparatus of the present invention is different from known application apparatus, and this is because it is suitable for allowing file system of the present invention carry out operational data to be transformed to file data.Application apparatus of the present invention can be taked the form of integrated circuit or be loaded into the form of the one or more executable files in the working storage of data handling system, and described data handling system is such as computing machine.
Application apparatus can be the machine that is specially a certain application-specific design, described application-specific such as computer integrated manufacturing system (CIM) equipment or Embedded Application.It preferably adopts the family that is equipped with application program or the form of office computer, and described application program comprises the Data Management Unit of mentioning just now.
Therefore, according to seventh aspect present invention, the present invention also comprises storage medium, this storage medium comprise in the mentioned executable file one of at least.Storage medium can be taked the form of disk, CD, Digital video disc or any other data storage medium such as fixed memory device.The present invention also provides a kind of permanent storage device, and permanent storage device comprises the code of representing storage management method of the present invention.
At last,, provide a kind of data handling system, comprised, according to file system device of the present invention, according to application apparatus of the present invention and/or according to storage medium of the present invention according to storage management apparatus of the present invention according to a eighth aspect of the present invention.
Below, will describe the present invention in detail according to preferred embodiment with reference to the accompanying drawings, wherein:
Fig. 1 a shows the non-volatile storage with operational data,
Fig. 1 b shows the non-volatile storage that has according to Fig. 1 a of the operational data of Fig. 1 a of file structure,
Fig. 1 c shows the non-volatile storage of Fig. 1 b, and wherein said non-volatile storage has the operational data of Fig. 1 b that is divided into file data blocks,
Fig. 1 d shows the disk of the file data blocks that comprises Fig. 1 c,
Fig. 2 to 7 shows non-volatile storage and disk in the different phase of preserving the file process according to the embodiment of the invention,
Fig. 8 a and b show during preserving the file process according to the embodiment of the invention, are included in the process flow diagram of message flow between application, memory manager program and the file system.
Fig. 1 a shows the non-volatile storage 10 that comprises by using the operational data structure of creating.Operational data is represented to L that by capitalization A it is included in four storage blocks 12 to 18.These storage blocks are linked by the pointer shown in arrow 20 to 24.The operational data of Fig. 1 a shows the structure that departs from the desired linear structure that comes from file.Storage block 12 by two pointers 20 and 22 parallel links to storage block 14 and 16.This parallel organization can not be represented in the file of the linear order that must comprise data.
Fig. 1 b shows the non-volatile storage 10 of Fig. 1 a.In Fig. 1 b, the operational data structure of Fig. 1 a is copied and is transformed to the operational data sequence that is included in the storage block 26.Reused storage block 12 to 18.For operational data structure 12 to 18 is transformed to file structure 26, used the conversion routine of concrete application, conversion routine is not shown in this Figure.
Fig. 1 c shows the non-volatile storage 10 with the file structure 26 that is divided into file data blocks 26.1 to 26.4.Each file data blocks comprises one group of data by letter representation again.For example, file data blocks 26.1 comprises data A, B and C.In addition, each data block is included in the distribution of sector on the disk 28, and it is shown in Fig. 1 d.Represent to distribute by " @ " symbol and the disk block address that is shown numeral in the figure.For example, the file data blocks 26.3 that comprises data G, H and I is distributed to file sectors 6.Carried out that by file system file data blocks is assigned to specific disk sector, described file system is not shown in this Figure.
Fig. 1 d shows the disk 28 with file sectors, and this document sector has file sectors address 1 to 9. File sectors 1,2,5 and 6 comprises data A to L.Situation about illustrating here is after the file data blocks 26.1 to 26.4 of Fig. 1 c being written to the disk sector that they distribute separately.Disk sector 3,4,7,8 and 9 temporary transient maintenances are idle so that distributed by file system.
Thereby Fig. 1 a shows being saved in the non-volatile storage 10 and the disk 28 of different phase of the process of disk 28 by being applied in the operational data of creating in the persistence working storage 10 to the sequence of 1d.Here the sequence of stages that illustrates often can be not different when using perishability storer rather than non-volatile storage 10.
Fig. 2 to 7 shows at file according to the embodiment of the invention and preserves the storage space 30 of persistent memory device of different phase of process and the disk space 32 of disc driver.Non-volatile storage 30 comprises two pieces 34 and 36, and it is a dot-hatched.The storage block of the dot-hatched in this figure and subsequent figures shows separately storage block under the control of file system, rather than under the control of memory manager program.With reference to figure 8 as described below deliver control according to the allocation step of the inventive method from memory manager program to the storage block of persistent-memory space 30.
Fig. 3 represents the stage of the back of identical systems.Because show identical system, so identical Reference numeral is used for the element of representing that those Fig. 2 are same.The second operational data structure 50 is write storage space.This operational data structure is the copy of operational data structure 38.Its data block 52,54 and 56 comprises the copy of operational data piece 40,42 and 44 respectively.Situation shown in Figure 3 was wherein used and is provided " being saved in disk " order corresponding to the stage that data is saved in the process of disk 32.The moment in " being saved in disk " order is written to data structure 50 to the state of operational data structure 38.
Fig. 4 is illustrated in the stage of the back of identical systems in the preservation process.In this stage, file system is handed in the control of the second operational data structure 50.This is shown by the storage block 52,54 and 56 of dot-hatched in Fig. 4.In addition, use and deliver conversion routine to file system, conversion routine is used for operational data structure 50 is transformed to file data structure.Conversion routine is stored in the storage block 58.Can select among the embodiment, read conversion routine and it is loaded into the piece 58 of non-volatile storage 30 from disk 32.Still provide original operational data structure in this stage.
Fig. 5 represents the stage of the back of this system.File system is applied to operational data structure 50 to conversion routine.Create file data structure like this in storage space 60, storage space 60 is distributed to file system by memory manager program.The memory range 38 that comprises original operational data piece 40,42 and 44 (referring to Fig. 2 to 4) has been removed from file system and be assigned to memory manager program.Therefore have the moment of the copy 50 of operational data structure 38 and conversion routine 58 when file system, operational data structure 39 has been safe this moment.Be exactly in this moment, file system is to using the success of confirming the preservation process.Memory manager program is freely distributed to Another Application or file system to this memory range.Obviously, partial memory scope 38 is distributed to file system so that store file data structure 60.
Fig. 6 represents the last stages of this system.In this stage shown here,, only file data structure 60 is kept in the persistent-memory space 30 according to the data relevant with carrying out the preservation process.Memory range 50 and piece 58 are removed distribution from memory manager program.Do not need them in the further process in current preservation process.
Fig. 7 is illustrated in the final stage of system in the preservation process.File data structure 60 is written to disk 32.Therefore, the subordinate storage block is removed distribution from memory manager program.Disk 32 present include files 60, file 60 is stored in four disk sectors 62.1 to 62.4.
Fig. 8 a and b according to the embodiment of the invention flow process there is shown execution method step and by use 70, memory manager program 72 and the operational data of file system 74 in the persistent storer be saved in the message that exchanges during the file process on the disk.Whole process spreads all over two Fig. 8 a and 8b.Just because the restriction of paper size just splits comes.
The preservation process of present embodiment by from use 70 to memory manager program 72 send hold-over command M10 and.So memory manager program will be by coming order is reacted for the copy memory allocated space of operational data in step S10, operational data will be stored on the disk.Then, in step S12, the copy of operational data is write the storage space that is distributed.In the present embodiment, carry out this step by memory manager program.Yet this step is preferably carried out by using 70.In further step S14, will distribute to file system to the storage space that comprises copying data (that is, according to the term that uses above " first of storage space ").Should be noted that memory manager program is by the request memory allocated space of using.This request is implicitly included among the message M10 that is preserved.
Then, give file system by message M12 distributing to report, message M12 also comprises the address realm of the storage space that is distributed, and wherein said address realm is shown by first address (Addr.1) and second address (Addr.2).File system acknowledges receipt of message M12 at step M14.
At step S16, memory manager program will keep memory allocation table and can not reallocate the storage space of distributing to file system to Another Application so that guarantee then.On the other hand, file system at step S18 with maintenance documentation allocation table (FAT) so that comprise the storage space that is distributed therein.
After sending message M10 whenever or when sending message M10, application will in message M16 be conversion routine from memory manager program request storage space, described conversion routine is transformed to file data to operational data.Memory manager program 72 will give be used memory allocation and report to using in message M18 at step S20, message M18 comprise from the address 3 to the address institute's addresses distributed scope of 4.At step S22, use conversion routine is written to the memory range that is distributed.
To distribute to file system to this memory range that comprises conversion routine at step S24 then.In message M20, report distribution to file system.File system 74 is confirmed to distribute in message M22.Then, at step S26 and S28, as described above, memory manager program maintenance memory allocation table and file system maintenance file allocation table.
In message M24, file system then will be to using the safety of confirming successfully to make operational data.
The further process of store method is described referring now to Fig. 8 b.In message M26, file system requests is used in the storage space of file data from the memory manager program branch, uses now the conversion routine under the control of file system to create described file data according to operational data.At step S30, memory manager program will distribute asked from the address 5 to the address 6 storage space, and correspondingly in message M28, report to file system.File system will be confirmed with message M30.Equally, memory manager program and file system will be safeguarded their allocation tables separately at step S32 and S34.
Then at step S36, file system initialization is converted to file data to the operational data 2 the storage space from address 1 to the address that is assigned to file system in S14.This makes file system for example wait for reasonable time according to working as the pre-treatment load at step S36.Will file data write from the address 5 to the address 6 the memory range that is distributed.At this moment wait, file system asks to redistribute storage space in message M32, and storage space comprises the copy (address 1 to the address 2) and the conversion routine (3 addresses 4, address) of operational data.Can select there is the independently message that is used for each storage space of mentioning among the embodiment.File system acknowledges receipt of the request redistributed and will safeguard its memory allocation table in step 38 in message M34.At the moment, mentioned memory space ranges is again under the control of memory manager program.File system is safeguarded its file allocation table at step S40.At step S42, file system initialization is written to disk to file data.File data is write disk comprise that the institute that shows with reference to figure 1c and d in steps.
When file data was stored into disk, file system was redistributed storage space from the memory manager program request in message M36, storage space include file data.Memory manager program is confirmed with message M38.Memory manager program and file system are safeguarded their allocation table at step S44 and S46.Use to writing the store method that cache memory is finished present embodiment by the persistent memory device.
Claims (13)
1. memory management apparatus that is used to manage the storage space of at least one persistent memory device, comprise storage allocation location, be used for communicating by letter with at least one application apparatus and being used at least one first of described storage space is distributed to described application apparatus, wherein said allocation units also are used for communicating by letter with at least one file system device, and are used for according to from described application apparatus or from the request of described file system device the first of described storage space being distributed to described file system.
2. memory management apparatus as claimed in claim 1, wherein said storage allocation location is used at current state maintenance memory allocation table, and described memory allocation table is given described application apparatus or described file system device the memory allocation of the described storage space part of at least one expression definition.
3. memory management apparatus as claimed in claim 2 also comprises processor and storer, wherein realizes described storage allocation location with the form that is included at least one first executable file in the described storer.
4. memory management apparatus as claimed in claim 3, wherein said storer is a persistent memory device, especially described persistent memory device.
5. file system device, comprise file allocation unit, be used at current state maintenance documentation allocation table, described file allocation table is distributed at least one file at least one disk space address, wherein said file allocation unit is used for communicating by letter with memory management apparatus, described memory management apparatus is relevant with persistent memory device, and is used for comprising in the described file allocation table of safeguarding the address of at least one first storage space of described persistent memory device.
6. file system device as claimed in claim 5 also comprises processor and storer, wherein realizes described file allocation unit with the form that is included at least one second executable file in the described storer.
7. application apparatus, comprise the persistent memory device that is connected to processor, with the Data Management Unit that is used for handling in the data of described permanent storage device, wherein said Data Management Unit is used for writing at least one the 3rd executable file to described permanent storage device, or be used for being provided to quoting at least one the 3rd executable file of described file system to described file system, thereby by carrying out described the 3rd executable file, it is predetermined data-sequence form that described processor is used for described data conversion.
8. application apparatus as claimed in claim 7, wherein storer particularly in described non-volatile storage the form with at least one the 4th executable file described Data Management Unit is provided.
9. storage medium comprises as claim 3,6,7 or 8 the first, second, third or the 4th executable file.
10. a data handling system comprises memory management apparatus as claimed in claim 1, file system device as claimed in claim 5, application apparatus as claimed in claim 7, or storage medium as claimed in claim 9.
11. a method that is used to manage the storage space of persistent memory device comprises according to from described file system device or at least one first of described storage space is distributed to the step of file system device from the request of application apparatus.
12. as the method for claim 11, wherein said allocation step comprises the step of obstruction to the described first write access of described storage space.
13. as the method for claim 12, wherein said allocation step comprises the ability of distributing the described first of the described storage space of read access to described file system device.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| EP02079656.1 | 2002-11-07 | ||
| EP02079656 | 2002-11-07 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN1879091A true CN1879091A (en) | 2006-12-13 |
Family
ID=32309412
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNA2003801028714A Pending CN1879091A (en) | 2002-11-07 | 2003-10-13 | Method and device for persistent-memory management |
Country Status (6)
| Country | Link |
|---|---|
| US (1) | US20060041731A1 (en) |
| EP (1) | EP1573550A2 (en) |
| JP (1) | JP2006518492A (en) |
| CN (1) | CN1879091A (en) |
| AU (1) | AU2003267777A1 (en) |
| WO (1) | WO2004042584A2 (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1797372B (en) * | 2004-12-23 | 2013-09-11 | 钟巨航 | Storing method and device in use for data process system |
| CN111868698A (en) * | 2018-03-20 | 2020-10-30 | 微软技术许可有限责任公司 | Free space direct connection |
Families Citing this family (27)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7930481B1 (en) | 2006-12-18 | 2011-04-19 | Symantec Operating Corporation | Controlling cached write operations to storage arrays |
| US9003104B2 (en) | 2011-02-15 | 2015-04-07 | Intelligent Intellectual Property Holdings 2 Llc | Systems and methods for a file-level cache |
| US9201677B2 (en) | 2011-05-23 | 2015-12-01 | Intelligent Intellectual Property Holdings 2 Llc | Managing data input/output operations |
| US8874823B2 (en) | 2011-02-15 | 2014-10-28 | Intellectual Property Holdings 2 Llc | Systems and methods for managing data input/output operations |
| US8996807B2 (en) | 2011-02-15 | 2015-03-31 | Intelligent Intellectual Property Holdings 2 Llc | Systems and methods for a multi-level cache |
| JP6552196B2 (en) * | 2011-08-02 | 2019-07-31 | アジャイ ジャドハブ | Cloud-based distributed persistence and cache data model |
| US9116812B2 (en) | 2012-01-27 | 2015-08-25 | Intelligent Intellectual Property Holdings 2 Llc | Systems and methods for a de-duplication cache |
| EP2845188B1 (en) | 2012-04-30 | 2017-02-01 | Nokia Technologies Oy | Evaluation of downbeats from a musical audio signal |
| CN102752374B (en) * | 2012-06-15 | 2015-04-29 | 中国电力科学研究院 | System and method for storing and accessing power utilization efficacy data |
| CN104620313B (en) | 2012-06-29 | 2017-08-08 | 诺基亚技术有限公司 | Audio signal analysis |
| US9612966B2 (en) | 2012-07-03 | 2017-04-04 | Sandisk Technologies Llc | Systems, methods and apparatus for a virtual machine cache |
| US10339056B2 (en) | 2012-07-03 | 2019-07-02 | Sandisk Technologies Llc | Systems, methods and apparatus for cache transfers |
| US9974476B2 (en) * | 2013-01-21 | 2018-05-22 | Innara Health, Inc. | Handheld cordless non-nutritive suck assessment device |
| US9842053B2 (en) | 2013-03-15 | 2017-12-12 | Sandisk Technologies Llc | Systems and methods for persistent cache logging |
| WO2015122924A1 (en) * | 2014-02-14 | 2015-08-20 | Hewlett-Packard Development Company, L.P. | Object based persistent main memory |
| WO2016118607A1 (en) * | 2015-01-20 | 2016-07-28 | Ultrata Llc | Distributed index for fault tolerant object memory fabric |
| WO2016118615A1 (en) | 2015-01-20 | 2016-07-28 | Ultrata Llc | Object memory data flow instruction execution |
| US9886210B2 (en) | 2015-06-09 | 2018-02-06 | Ultrata, Llc | Infinite memory fabric hardware implementation with router |
| US9971542B2 (en) | 2015-06-09 | 2018-05-15 | Ultrata, Llc | Infinite memory fabric streams and APIs |
| US10698628B2 (en) | 2015-06-09 | 2020-06-30 | Ultrata, Llc | Infinite memory fabric hardware implementation with memory |
| US10235063B2 (en) | 2015-12-08 | 2019-03-19 | Ultrata, Llc | Memory fabric operations and coherency using fault tolerant objects |
| US10241676B2 (en) | 2015-12-08 | 2019-03-26 | Ultrata, Llc | Memory fabric software implementation |
| WO2017100281A1 (en) | 2015-12-08 | 2017-06-15 | Ultrata, Llc | Memory fabric software implementation |
| EP3387548B1 (en) | 2015-12-08 | 2023-08-02 | Ultrata LLC | Memory fabric operations and coherency using fault tolerant objects |
| US10402101B2 (en) | 2016-01-07 | 2019-09-03 | Red Hat, Inc. | System and method for using persistent memory to accelerate write performance |
| US20180004649A1 (en) * | 2016-07-01 | 2018-01-04 | Intel Corporation | Techniques to Format a Persistent Memory File |
| US10657068B2 (en) * | 2018-03-22 | 2020-05-19 | Intel Corporation | Techniques for an all persistent memory file system |
Family Cites Families (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| AU662973B2 (en) * | 1992-03-09 | 1995-09-21 | Auspex Systems, Inc. | High-performance non-volatile ram protected write cache accelerator system |
| JPH0962560A (en) * | 1995-08-30 | 1997-03-07 | Fuji Xerox Co Ltd | Database device |
| US5930167A (en) * | 1997-07-30 | 1999-07-27 | Sandisk Corporation | Multi-state non-volatile flash memory capable of being its own two state write cache |
| US6480935B1 (en) * | 1999-01-15 | 2002-11-12 | Todd Carper | Smart card memory management system and method |
| EP1498810B1 (en) * | 1999-10-21 | 2018-12-26 | Panasonic Corporation | A semiconductor memory card access apparatus, a computer-readable recording medium, an initialization method, and a semiconductor memory card |
| US6804763B1 (en) * | 2000-10-17 | 2004-10-12 | Igt | High performance battery backed ram interface |
| US20030182414A1 (en) * | 2003-05-13 | 2003-09-25 | O'neill Patrick J. | System and method for updating and distributing information |
| US20030004922A1 (en) * | 2001-06-27 | 2003-01-02 | Ontrack Data International, Inc. | System and method for data management |
| US20030081932A1 (en) * | 2001-10-25 | 2003-05-01 | Hanes David H. | Computer-readable medium and method for providing a generic interface to a CD-recorder device |
| US6901499B2 (en) * | 2002-02-27 | 2005-05-31 | Microsoft Corp. | System and method for tracking data stored in a flash memory device |
| US20030212865A1 (en) * | 2002-05-08 | 2003-11-13 | Hicken Michael S. | Method and apparatus for flushing write cache data |
-
2003
- 2003-10-13 WO PCT/IB2003/004564 patent/WO2004042584A2/en active Application Filing
- 2003-10-13 EP EP03748472A patent/EP1573550A2/en not_active Withdrawn
- 2003-10-13 US US10/533,735 patent/US20060041731A1/en not_active Abandoned
- 2003-10-13 JP JP2004549421A patent/JP2006518492A/en active Pending
- 2003-10-13 CN CNA2003801028714A patent/CN1879091A/en active Pending
- 2003-10-13 AU AU2003267777A patent/AU2003267777A1/en not_active Abandoned
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1797372B (en) * | 2004-12-23 | 2013-09-11 | 钟巨航 | Storing method and device in use for data process system |
| CN111868698A (en) * | 2018-03-20 | 2020-10-30 | 微软技术许可有限责任公司 | Free space direct connection |
Also Published As
| Publication number | Publication date |
|---|---|
| AU2003267777A1 (en) | 2004-06-07 |
| WO2004042584A3 (en) | 2006-06-22 |
| JP2006518492A (en) | 2006-08-10 |
| AU2003267777A8 (en) | 2004-06-07 |
| EP1573550A2 (en) | 2005-09-14 |
| US20060041731A1 (en) | 2006-02-23 |
| WO2004042584A2 (en) | 2004-05-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN1879091A (en) | Method and device for persistent-memory management | |
| AU2012352178B2 (en) | Working set swapping using a sequentially ordered swap file | |
| US9436597B1 (en) | Using non-volatile memory resources to enable a virtual buffer pool for a database application | |
| US10747673B2 (en) | System and method for facilitating cluster-level cache and memory space | |
| RU2509347C2 (en) | Flexible enlargement of memory pages | |
| JP2019020788A (en) | Memory system and control method | |
| US10019363B2 (en) | Persistent memory versioning and merging | |
| US20080059728A1 (en) | Systems and methods for masking latency of memory reorganization work in a compressed memory system | |
| KR20100132244A (en) | Memory system and method of managing memory system | |
| US11132130B2 (en) | Segment cleaning method using non-volatile random-access memory and memory management apparatus thereof | |
| WO2018093442A1 (en) | Storage operation queue | |
| JP2008299848A (en) | Data processing apparatus and method for nonvolatile memory | |
| TW201222249A (en) | Storage device | |
| US20130346714A1 (en) | Hardware-Based Accelerator For Managing Copy-On-Write | |
| CN1637725A (en) | Method, system, and article of manufacture for reserving memory | |
| KR20180044095A (en) | Electronic systems and methods of operating thereof | |
| CN1506843A (en) | Data processing system capable of using virtual memory processing mode | |
| US10140029B2 (en) | Method and apparatus for adaptively managing data in a memory based file system | |
| JP2003256269A (en) | Control method for nonvolatile storage device, and memory device | |
| CN102160038A (en) | Method and an apparatus to manage non-volatile disl cache | |
| JP6254986B2 (en) | Information processing apparatus, access controller, and information processing method | |
| US5430858A (en) | Method for RAM conservation employing a RAM disk area non-sequential addresses on arranged order basis to access executable procedures | |
| JP4792065B2 (en) | Data storage method | |
| CN113448722A (en) | Mapping method of process memory and instance processing method based on serverless architecture | |
| JP7337228B2 (en) | Memory system and control method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| ASS | Succession or assignment of patent right |
Owner name: NXP CO., LTD. Free format text: FORMER OWNER: KONINKLIJKE PHILIPS ELECTRONICS N.V. Effective date: 20071102 |
|
| C41 | Transfer of patent application or patent right or utility model | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20071102 Address after: Holland Ian Deho Finn Applicant after: Koninkl Philips Electronics NV Address before: Holland Ian Deho Finn Applicant before: Koninklijke Philips Electronics N.V. |
|
| C12 | Rejection of a patent application after its publication | ||
| RJ01 | Rejection of invention patent application after publication |
Open date: 20061213 |